60% of the sub-Saharan African population and 85% of Ugandans reside in rural areas, where there is an urgent need to increase coverage of male circumcision (MC). The Rakai Health Sciences Program (RHSP) has conducted randomized trials of MC for HIV/STI prevention, implementation science research on PEPFAR- funded MC services, and has monitored MC uptake and effectiveness in a rural population of 14,000 persons enrolled in a 50 village Rakai Community Cohort Study (RCCS). Between 2007-mid 2011, we performed >18,000 service MCs in 5 static facilities; recently, we also piloted MC provision in a mobile MC camp. However, MC coverage among non-Muslim men in the RCCS population remains modest at ~28%, and MC acceptance occurs disproportionately among younger, lower risk men. We propose the following implementation science research. Aim 1. To increase MC uptake in a rural setting, particularly in higher risk men, using enhanced messaging, targeted outreach, task shifting, task sharing and mobile camps, and to compare MC uptake, coverage, and client sociodemographic/behavioral characteristics and cost-effectiveness in static clinics versus mobile camps. We project ~27,000 MC acceptors overall within 3 years. We hypothesize that the enhanced messaging/outreach will increase overall MC uptake by higher risk men compared to our prior services, that mobile MC camps will increase acceptance by higher risk men and be more cost-effective than static services in this rural setting, and that MC prevalence in RCCS will increase to 60%. Aim 2. Early infant MC (EIMC) is simpler, safer and cheaper than adult MC, and can have a long-term impact on the HIV epidemic. Sustainable scale-up of EIMC requires task shifting to non-physicians. We will conduct a randomized study of the acceptability and safety of EIMC provided by clinical officers (COs) who currently perform adult MC, compared with nurse-midwives (NMWs) who are not currently licensed to perform the procedure. We will first train personnel in EIMC using the Mogen clamp under local anesthesia, and then enroll 500 infants aged <2 months per arm. We hypothesize that the acceptability and safety of EIMC performed by NMWs will be comparable to COs. Outcomes will be shared with policy makers for future program development. Aim 3. To assess the population prevalence of MC and its impact on HIV incidence in the RCCS population (supported almost completely by other grants), and estimate the number of MCs required to avert one HIV infection and costs per HIV infection averted, using marginal structural models for causal inference. A stochastic simulation model will be used to project infections averted and cost savings over 10 years. The proposed studies will provide a model for rapid MC scale up (Aim 1), and EIMC services (Aim 2) to help guide programs in rural Africa. The impact of MC on HIV incidence, cost-effectiveness and enhanced modeling (Aim 3) will provide data on population prevention of HIV through MC, for policy and program development.